It is recognized that high-speed manufacture helps to drive down the cost of a broad range of products ranging from consumable items and materials to electronic and optical components. Web-based fabrication, conventionally used for making photographic film and related sensitized materials, is particularly advantaged for high-speed manufacture of products formed on flexible substrates. Thus, methods that utilize web-based fabrication enable more economical manufacturing solutions for existing products and enable the development of new products on flexible substrates.
One area of particular interest for high-speed, web-based manufacture relates to the fabrication of electronic or optical devices on flexible substrates. Methods for device fabrication would form the component elements of such electronic or optical devices by depositing patterned layers of material, in liquid droplet form, onto a substrate. In web-based printing, a flexible medium, typically provided in roll or cut sheet form, is fed to the printing mechanism and is moved through the printing mechanism during application of inks or other materials. In concept, web-based fabrication would adapt this printing model for manufacturing electronic and other devices on a flexible support or substrate. In order to fabricate electronic or optical devices commercially onto a moving flexible substrate using liquid droplet deposition, the following requirements are of special importance:                (i) High resolution (dots per inch). The capability for fine detail needed to form layers within an electronic component places high demands on droplet resolution, that is, on spatially compact and accurate placement of droplets. Although high-resolution can be achieved by executing multiple passes of the receiving surface past a printing mechanism, it would be highly preferable to deposit all droplets of a single fluid from the same droplet forming or droplet ejection station in one pass to form a structure. This cannot be done with conventional drop on demand ink jet technology because such technology cannot apply liquids with laydown densities sufficient to form many desirable structural features in a single pass. Accordingly, multiple passes are required to form such structures using conventional droplet demand ink to technology. This creates the potential for registration errors which can affect the quality of the formed structure. Further, such an approach creates striations within the formed structure that can alter the performance capabilities of the structure. For example, in FIG. 1A, a cross-section view of a structure 3 is shown that is formed using such prior art technologies. As is shown in FIG. 1A, a first layer 4 of structure 3 is formed on a substrate 2 in a first printing pass. Similarly, second layer 5 is applied to first layer 4 in a second pass, while a third layer 6 is applied in a third pass. It will be appreciated that the nature of and quality of the material boundaries between first layer 4, second layer 5, and third layer 6 limits the functional capabilities of structure 3 so formed. For example these boundaries, referred to herein as striations can alter the conductance of structure 3, the mechanical strength of structure 3, the thermal management capability to the structure 3 or other desirable properties of structure 3. Accordingly, what is desired is a system that is capable of forming structures comprising a relatively large feature height but with a minimum number of striations. Undesirable artifacts may be visible in a top view as shown in FIG. 1C, which is a top view the structure illustrated in section in FIG. 1A. Such artifacts can be caused by deposition of discrete drops too widely separated in time which, in turn, causes registration problems.        (ii) Dependable delivery of fluid droplets at the needed volume and flow rate, without interruption. Related both to resolution and to yield percentages for a continuously moving substrate, this requirement concerns the overall speed, reliability, and robustness of the droplet-forming mechanism itself. Since web fabrication uses a continuously moving substrate, there is no opportunity to rescan an area of the substrate to ensure droplet delivery once the substrate has moved past the droplet-ejection station. In a single pass, it would be preferable to be able to form a feature on a surface where the height of the feature is relatively large. The relative height of a feature can be best expressed in terms of a height:width ratio for the feature, as formed by the droplet ejection mechanism in a single pass. A height:width ratio of at least 0.20 must be achievable with the substrate traveling at a rate of speed exceeding a few inches per second.        (iii) Highly accurate droplet positioning. Also related to resolution and yield, the positional accuracy of the droplet-ejection mechanism must allow droplet placement accuracy preferably within +/−2 microns. For a straight line feature extending in any direction along the flexible substrate, height of the feature should be controllable to within about 1% RMS. Width should be controllable to within about 1% RMS.        (iv) Precision control of droplet volume. Similarly related to resolution and yield, the stringent requirement for droplet volume control assures that precisely the correct amount of fluidic material is deposited at each location on the flexible substrate surface. For electronic component fabrication, droplet volumes would range from about 3 to about 40 picoliters, with volume controlled to within about +/−1% or better. The capability to change droplet volume and maintain this tight level of control would be particularly advantageous for component fabrication.        (v) Flexibility for droplet treatment, in flight and on impact. For some types of fluid materials, it may be advantageous to provide some type of conditioning treatment, such as to obtain proper adhesion, for example, or to prevent or facilitate rapid drying. This function can be most easily achieved where the distance allowable between the droplet ejection mechanism and the substrate surface is somewhat flexible. For a broad range of applications, it would be advantageous to allow a gap between the ejection point of a liquid droplet and the substrate surface.        (vi) Capability to handle a range of fluid viscosities. The many types of fluids that must be deposited for forming electronic components and other complex devices span a wide range of viscosities. High-viscosity fluids, in particular, such as those with a low shear in excess of 10 cP, can prove difficult to deposit accurately in droplet form. A number of fluid materials with advantageous properties for forming electronic devices have low shear viscosities in excess of 100 cP, well beyond the range of conventional droplet deposition devices. One problem inherent to droplet-based delivery of high viscosity fluids relates to undesirable satellite droplets that form as a result of droplet formation. Generally tinier than printing droplets, satellite droplets form due to the complex rheology of these substances.        
FIG. 1B shows a top view of a substrate 2 having intentionally printed drops 7 of material adjacent to each other with undesirable satellite drops 8 and 9 in the vicinity of the intentionally printed drops 7. Such satellite drops can have a number of undesirable effects on the structure formed on substrate 2. For example, as is shown in FIG. 1B, satellite drop 8 is positioned proximate to one of the intentionally printed drops 7 so as to effectively alter the shape thereof. As is also shown in FIG. 1B, satellite drops 9 are positioned between intentionally printed drops 7 and can, in certain embodiments, provide an undesirable electrical short therebetween or create other unintentional effects.                (vii) Adaptability to a range of fluid characteristics. In addition to viscosity, there are a number of other fluid properties that present special challenges for deposition onto a flexible substrate in droplet form. Volatile and other heat-sensitive fluids must be carefully handled and may not be suitable for droplet ejection methods that employ high heat levels. Highly viscous polymers can be particularly sensitive to high heat levels. Fluids that exhibit electroluminescence must also be protected from high heat conditions. The droplet formation and ejection mechanisms would ideally be capable of depositing both conductive and non-conductive fluids. It would be advantageous to be able to form and eject fluid droplets containing colloids and particulates, including fluids containing suspended nanoparticles. In addition, it would be beneficial to be able to adapt the deposition mechanism to the rheology of the deposited fluid, rather than being constrained by inherent limitations of the fluid delivery technology.        (viii) Suitable surface preparation. The method adapted for fluid deposition must be compatible with various types of surface preparation techniques for improved droplet adhesion, spread, and related characteristics.        
Conformance to the above-listed requirements is well beyond the capabilities of conventional methods for droplet deposition. The ability to meet or exceed these requirements would enable web-based fabrication with increasingly faster throughput, possibly allowing web media transport speeds in excess of 1000 feet per minute for some types of applications and components.
Conventional techniques for the fabrication of electronic and electro-optical devices typically involve a number of different processes for forming various layers that make up the device, including photolithography, oxidation, etching, and masking, for example. Techniques for depositing materials in a droplet or vaporized form to build surface features include vacuum or vapor deposition, sputtering, and droplet deposition using spray bar apparatus. Often, multiple processes are used in combination, requiring transfer of a substrate between various types of equipment and involving careful handling of the product in its intermediate fabrication stages. Because of this, the complex processing sequence that is currently required for fabrication of a display device or for a support electronic component such as a field effect transistor is time-consuming, trouble-prone, and costly. As the number of processing steps increases, the technical challenges for integration of components on a flexible substrate become even more demanding, throughput slows, the likelihood of contamination increases, and yields can be dramatically reduced.
Attempts to improve conventional fabrication techniques and achieve more satisfactory yields have included use of drop-on-demand ink jet print heads for deposition of at least some layers of electronic or optical components, as is disclosed, for example, in U.S. Pat. No. 6,503,831 to Speakman; U.S. Pat. Nos. 6,194,837 and 6,545,424 to Ozawa; U.S. Pat. Nos. 6,373,453 and 6,642,651 to Yudasaka; U.S. Pat. No. 6,555,968 to Yamazaki et al.; and U.S. Pat. No. 6,087,196 to Sturm et al.
In operation, “drop-on-demand” ink jet printing provides fluid droplets for impact upon a recording surface using a localized pressurization actuator (thermal, piezoelectric, air pressure, etc.) at each nozzle. Selective activation of the actuator causes the formation and ejection of a droplet from a corresponding nozzle. The droplet crosses the space between the print head nozzle and the print substrate and strikes the print substrate. The formation of printed images, for example, is achieved by controlling the individual formation of ink droplets, as is required to create the desired image. With thermal actuators, a heater for each nozzle, located at a convenient location, heats the fluid within a chamber, causing a quantity of the fluid to change phase and to form a gaseous steam bubble. This momentarily increases the internal fluid pressure sufficiently for a fluid droplet to be expelled. The bubble then collapses as the heating element cools, and the resulting vacuum draws fluid into the chamber from a reservoir to replace fluid that was ejected from the nozzle. Alternately, piezoelectric actuators, such as that disclosed in U.S. Pat. No. 5,224,843 to vanLintel, have a piezoelectric crystal in a fluid channel that flexes when an electric current flows through it, forcing a fluid droplet out of a nozzle.
In conventional thin-film fabrication methods, such as those used in commercially available drop-on-demand ink jet printers, a sheet of substrate is held stationary during materials deposition. One or more print heads or other printing mechanisms are then passed over the stationary substrate, one or more times, in order to deposit the various component layers with sufficient resolution to form the electronic device. Once deposition of these materials is complete, the sheet of substrate can be lifted from place and made available for any further processing.
However, while drop-on-demand ink jet print mechanisms have been adapted for depositing some types of materials onto a substrate, there are significant drawbacks and performance thresholds, inherent to drop-on-demand technology, that limit its usefulness for web-based fabrication of electronic and related components, in which the substrate is continuously moving. The resolution limitations of drop-on-demand printers are a function of the droplet formation and delivery, componentry design, with a separate heater or piezoelectric actuator required for each individual nozzle. Characteristically, drop-on-demand printers improve their inherently low resolution by making a series of repeated passes over the same area of a substrate. The use of repeated passes, however, would not be well-suited to a web manufacturing environment with a continuously moving substrate. In addition, layering of the same material onto itself in successive passes is not optimal for obtaining homogeneous density. For most deposited materials, striations develop at the interface between successively printed layers, which is undesirable in many applications, for example in the deposition of Organic Light Emitting Diode (OLED) or Polymeric Light Emitting Diode (PLED) materials.
Droplet volume control is typically well above +/−5% at best, a limitation that is further compounded by tendency of nozzles to clog or to form and eject unwanted satellite droplets. There is no known method for compensating for the formation of satellite droplets that can deposit material improperly, compromising component operation or even causing component failure. Satellite droplets are commonly observed with drop-on-demand print devices, particularly where fluids deposited exceed viscosities of about 3 cP. The viscosity range of drop-on-demand printers is very limited; high viscosity fluids can exhibit low-shear viscosities well out of range of accurate droplet formation and delivery for these devices. Because many types of drop-on-demand print heads employ pulsed heat for forming and ejecting droplets, there are also limitations on the types of fluids that can be accurately delivered without being damaged or introducing safety problems. Requiring a close proximity between the nozzle and receiving substrate, the drop-on-demand print head is relatively inflexible for allowing supporting droplet conditioning mechanisms to be used. Drop-on-demand print mechanisms cannot be scaled, beyond a narrow range, to suit the rheology of the fluid to be deposited; instead, the fluid must be adapted to the fairly limited geometry of the drop-on-demand droplet forming mechanism.
To counter some of the inherent limitations of this technology, some types of drop-on-demand solutions use a wax carrier into which a colorant or dye is mixed. For this type of print technology, the wax carrier remains as part of the deposited material. While this may be acceptable for some types of color printing, it can be readily appreciated that the retention of a wax carrier would prevent component fabrication in most cases. Finally, while the quality of patterns formed using drop-on-demand ink jet techniques can be acceptable at lower throughput rates where the substrate can be held stationary, the demands of web fabrication at high rates of speed easily exceed the capabilities of drop-on-demand technology and restrict the types of components that can be fabricated or require combination with other types of deposition methods to supplement the droplet formation provided from the print head.
Recognizing the shortcomings of conventional inkjet printing for component fabrication, a number of hybrid methods have been proposed. These include methods disclosed in the Ozawa '837 and '424 patents cited above, in which, for a device comprising multiple layers of material, some of the component layers are deposited using drop-on-demand ink jet printing and other layers are deposited using vacuum deposition or other methods, including conventional use of masks and photochemical etching, as is described in International Application WO 97/18944 by Calvert et al.; European Patent EP 0 615 256 to Mutsaers et al.; European Patent Application EP 1 079 397 to Cloots et al.; and U.S. Pat. No. 5,976,284 to Calvert et al. However, it can be appreciated that such hybrid methods are not particularly conducive to high-speed web-based fabrication and add cost and complexity to the fabrication process. As is true with conventional photo-etching schemes, the substrate itself must be maintained in place during the deposition process, effectively precluding any type of web-based fabrication. Given the constraints inherent to conventional drop-on-demand ink jet print apparatus, it would be difficult to further extend the use of drop-on-demand ink jet masking technology for forming conductive patterns of higher complexity and high resolution and to obtain the economies of high-speed production afforded by web-based fabrication.
Increasing demand for economical, high-speed methods of fabrication for electronic, electro-optical, or optical devices, particularly for Organic Light Emitting Diodes (OLED) and Polymeric Light Emitting Diodes (PLED) sometimes known as solution processable organic light emitting diode display components, points to the need for improvements over the conventional fabrication methods that have been commercialized to date. Thus, it can be seen that there is a need for an improved method and apparatus for forming two- and three-dimensional structures on flexible substrates using web-based fabrication techniques, in which the substrate is continuously moving in a travel direction.